Abstract--Recently, sparse approximation has become a preferred method for learning large scale kernel machines. This technique attempts to represent the solution with only a subse...
Soft computing (SC) emerged as an integrating framework for a number of techniques that could complement one another quite well (artificial neural networks, fuzzy systems, evolutio...
We propose a multiple incremental decremental algorithm of Support Vector Machine (SVM). Conventional single incremental decremental SVM can update the trained model efficiently w...
Statistical learning methods are emerging as a valuable tool for decoding information from neural imaging data. The noisy signal and the limited number of training patterns that ar...
We consider the problem of semi-supervised feature selection, where we are given a small amount of labeled examples and a large amount of unlabeled examples. Since a small number ...